Dietary Factors and Chronic Kidney Disease

The role of diet in maintaining a healthy weight in decreasing the risk for diabetes and hypertension—the leading causes of kidney failure worldwide—is undoubtedly important. Recent evidence suggests that dietary factors may also directly influence decline in kidney function. Nutritional epidemiology has traditionally focused on the development of diabetes and cardiovascular disease. Logically, the same dietary factors implicated in macrovascular coronary or cerebral vascular disease will also manifest in microvascular disease of the kidneys.

Several challenges arise in studying the role of diet in kidney disease progression in humans. First, as in any investigation of dietary factors and the development or progression of chronic medical conditions, adherence in an interventional diet study is difficult to maintain in randomized participants over several weeks or years. Second, kidney disease progression usually requires several years to manifest in community-dwelling adults in the general population. So the majority of studies with the requisite long-term follow-up in this area are observational and subject to potential confounding by unmeasured entities that may reflect an overall healthier lifestyle. Nonetheless, longitudinal observational studies currently provide the majority of information for associations between diet and kidney disease.

In critical evaluation of the published medical literature, additional considerations such as how kidney disease progression is defined and how diet is administered or measured are also important. Here we summarize the major relevant research studies and divide “kidney disease” into two main entities: 1) directly measured GFR or estimated GFR (eGFR), widely considered to be the primary measure of kidney “function” (Table 1), and 2) the presence of microalbuminuria, which is commonly considered to represent early kidney disease as well as reflect systemic vascular dysfunction (Table 2). A variety of kidney outcomes have been examined by different investigators, which may make direct comparisons between studies difficult. Notably, almost all published studies looking at microalbuminuria are cross-sectional.


How dietary intake is measured and quantified in observational studies also deserves attention. In longitudinal cohort studies, a common evaluation tool is the semi-quantitative food frequency questionnaire, which assesses average food intake over the preceding year in approximately 130 items. Responders are given a standard portion size and choose one of nine possible frequency-of-consumption responses, ranging from “never or less than once per month” to “six or more times per day” for each food item. Total energy and nutrient intake can then be calculated by summing up energy or nutrients from all foods. Whereas traditional nutritional epidemiology has focused on individual nutrients or foods, their additive or interactive influence perhaps may be better observed when overall diet patterns are considered for incident chronic diseases. Therefore, nutritional epidemiology studies in recent years have included analyses of healthful dietary patterns (e.g., prudent-style and Dietary Approaches to Stop Hypertension [DASH]-style, both high in whole grain, fruit, and vegetable intake) and of unhealthful dietary patterns (e.g., Western-style, high in red meat, refined grains, and sweets).

Historically, the role of dietary protein in kidney disease has been dominant because of a number of longstanding reports that protein restriction delays the progression of kidney function decline in laboratory animals. In 1994, results were published from the Modification of Diet in Renal Disease study, a multicenter randomized controlled trial of 840 adults with chronic kidney disease (GFR 25–55 mL/min per 1.73 m2) who were randomized to usual-protein (1.3 g/kg per day), low-protein (0.58 g/kg per day), or very-low-protein (0.28 g/kg per day) diets. The primary finding was that there was no significant difference between the diet groups in mean GFR decline over 2.2 years. Limitations of this study include the short follow-up time and the exclusion of all but diet-controlled diabetic participants. By contrast, subsequent analyses of the Nurses’ Health Study observational cohort by Knight et al. (Table 1) reported that higher dietary animal (but not total, dairy, or vegetable) protein intake was associated with faster eGFR decline over 11 years in women (only ∼4% diabetic) with baseline mild renal insufficiency (defined as eGFR 55–80 mL/min per 1.73 m2). The Modification of Diet in Renal Disease study did not distinguish between different types of dietary protein, which may have differential effects on eGFR decline.

Over recent years, additional studies on dietary factors and chronic disease have been published in the Nurses’ Health Study, the Multi-Ethnic Study of Atherosclerosis, the Atherosclerosis Risk in Communities, and the Reason for Geographic and Racial Differences in Stroke cohorts (Tables 1 and 2). Overall, the majority of the findings suggest that diets considered “heart healthful” (low in saturated animal fats and protein, sodium, and sweetened beverages but high in fruit, vegetables, high-fiber whole grains, low-fat dairy, and fish) are inversely associated with the presence and progression of chronic kidney disease.

A potential pathophysiologic link between diet and kidney disease (as well as heart disease) is inflammation. Interestingly, inflammatory markers such as C-reactive protein, intracellular adhesion molecule-1, and vascular cellular adhesion molecule-1—which have been associated with subsequent coronary heart disease, the presence of albuminuria, and faster decline of kidney function in multiple studies—are significantly more elevated in people eating unhealthful diets than those eating healthful diets. More research is needed, however, to further assess the role of diet in modifying the risk for chronic kidney disease progression.


[1] Amaka Eneanya, MD, (nephrology fellow) and Julie Lin, MD, MPH, FASN, (faculty physician) are members of the Renal Division, Department of Medicine at Brigham and Women’s Hospital and Harvard Medical School in Boston.

October-November 2011 (Vol. 3, Number 10 & 11)